Inhibitor of �B Kinase � (IKK�), STAT1, and IFIT2 ProteinsDefine Novel Innate Immune Effector Pathway against WestNile Virus Infection*□S

Received for publication, July 26, 2011, and in revised form, November 4, 2011 Published, JBC Papers in Press, November 7, 2011, DOI 10.1074/jbc.M111.285205

Olivia Perwitasari‡§, Hyelim Cho¶, Michael S. Diamond¶�**, and Michael Gale, Jr.‡§1

From the ‡Department of Immunology and §Molecular and Cellular Biology Program, University of Washington School ofMedicine, Seattle, Washington 98195 and the Departments of ¶Molecular Microbiology, �Pathology and Immunology, and**Medicine, Washington University School of Medicine, St. Louis, Missouri 63110

Background: IFN activates JAK-STAT signaling, where STAT1phosphorylation is crucial for ISG induction and expressionof IFIT2 to limit West Nile virus infection.Results: IKK� mediates STAT1 serine 708 phosphorylation exclusive of tyrosine phosphorylation but dependent on nuclearexport and ISG synthesis.Conclusion: IKK�-mediated STAT1 S708 phosphorylation is crucial for IFIT2 expression to control WNV.Significance:We define a novel anti-WNV innate immune effector pathway.

West Nile virus is an emerging virus whose virulence is depen-dent upon viral evasion of IFN and innate immune defenses. Theactions of IFN-stimulated genes (ISGs) impart control of virusinfection, but the specific ISGs and regulatory pathways thatrestrictWestNile virus (WNV) are not defined.Herewe show thatinhibitorof�Bkinase� (IKK�)phosphorylationofSTAT1atserine708 (Ser-708) drives IFIT2 expression tomediate anti-WNVeffec-tor function of IFN. WNV infection was enhanced in cells fromIKK��/� or IFIT2�/� mice. In IKK��/� cells, the loss of IFN-in-duced IFIT2 expressionwas linked to lack of STAT1phosphoryla-tiononSer-708butnotTyr-701norSer-727.STAT1Ser-708phos-phorylation occurs independently of IRF-3 but requires signalingthrough the IFN-�/� receptor as a late event in the IFN-inducedinnate immuneresponsethatcoincideswithIKK�-responsiveISGsexpression. Biochemical analyses show that STAT1 tyrosinedephosphorylation and CRM1-mediated STAT1 nuclear-cyto-plasmic shuttling are required for STAT1 Ser-708 phosphoryla-tion. When compared with WT mice, WNV-infected IKK��/�

mice exhibit enhanced kinetics of virus dissemination andincreased pathogenesis concomitant with loss of STAT1 Ser-708phosphorylation and IFIT2 expression. Our results define an IFN-induced IKK� signaling pathway of specific STAT1 phosphoryla-tion and IFIT2 expression that imparts innate antiviral immunityto restrictWNV infection and control viral pathogenesis.

West Nile virus (WNV)2 is an emerging flavivirus that hasrecently spread into the Western hemisphere from points of

origin within Asia, Africa, or theMiddle East (1, 2). Infection byWNV is now a leading cause of arboviral encephalitis andimparts 4% overall case fatality frequency in the United States.Typicallymaintainedwithin avian reservoirs,WNV is spread toother vertebrates, including humans as dead-end hosts,through mosquito bite. WNV circulates as two major lineagesand minor clades, with specific clades of lineage 1 representingthe emergent and virulent strain in North America and else-where, whereas lineage 2 strains are typically endemic to Africaand Asia and are not known to cause disease in humans (1,3–6). WNV infection is controlled in part through type I IFNimmune defenses (7, 8). IFN actions comprise a major compo-nent of the innate immune response to virus infection, whichserves to restrict virus replication and spread in part throughthe actions of interferon-stimulated genes (ISGs). WNV sup-pression of IFN signaling is linked to virus dissemination, neu-roinvasion, and virulence of emergent lineage 1 strains (8, 9).WNV acutely induces IFN-� expression from infected cells

upon engagement of RNAproducts, including viral RNAby theRIG-I-like receptors, RIG-I andMDA5 (10–12). TheRIG-I-likereceptors signal the downstream activation of interferon regu-latory factor (IRF) 3 and NF-�B transcription factors throughthe actions of the IKK-related kinases (Tank binding kinase 1(TBK1) and inhibitor of �B kinase � (IKK�)). As a result, IRF-3and NF-�B activation drive the expression of IFN-�, other pro-inflammatory cytokines, chemokines, and direct IRF-3-targetgenes that confer antiviral and immune-activating functions(13, 14). Secreted IFN-� drives the innate immune responsecharacterized by ISG expression. This process is triggered uponIFN-� binding to the interferon receptor (IFNAR) to inducereceptor dimerization, autophosphorylation of receptor-asso-ciated kinases Tyk2 and JAK1 leading to tyrosine phosphoryla-tion of signal transducer and activator of transcription (STAT)1

* This work was supported, in whole or in part, by National Institutes of Healthgrants R01AI074973 and U19AI083019.

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. S1–S6.

1 To whom correspondence should be addressed: University of Washing-ton School of Medicine, Department of Immunology, Box 357650, Seat-tle, WA 98195. Tel.: 206-685-7953; Fax: 206-643-1013; E-mail: mgale@u.washington.edu.

2 The abbreviations used are: WNV, West Nile virus; ADAR, double-strandedRNA-specific adenosine deaminase; IFNAR, interferon-�/� receptor; ISG,

interferon-stimulated gene; IRF, interferon regulatory factor; IKK, I�Bkinase; SenV, Sendai virus; m.o.i., multiplicity of infection; CHX, cyclohexi-mide; LMB, leptomycin B; MEF, mouse embryonic fibroblast; ISRE, interfer-on- stimulated regulatory element; PKR, protein kinase R.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 286, NO. 52, pp. 44412–44423, December 30, 2011© 2011 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

44412 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 52 • DECEMBER 30, 2011

at residueTyr-701, phosphorylation of STAT2, and assembly ofthe STAT1-STAT2-IRF9 ISGF3 complex (15–18). ISGF3 func-tion is further augmented or sustained by STAT1 serine phos-phorylation at residues Ser-727 and Ser-708 (19–21). ISGF3translocates to the nucleus and triggers the transcription ofhundreds of ISGs (13, 15). ISG products serve as immuno-modulators and restriction factors against virus infection(22–24).Among the ISGs, IFN-induced protein with tetratricopep-

tide repeats (IFIT)2, also known as ISG54, has been identified asan ISG restriction factor of WNV (23). IFIT2 belongs to theIFIT gene family whose members function to restrict virusinfection through alteration of cellular protein synthesis(reviewed in Ref. 25), and IFIT2mediates these action by inhib-iting eIF3 function in translation initiation (26–28). Our recentstudy revealed that in the absence of IFN, ectopic IFIT2 expres-sion can impose a blockade that ultimately restricts WNV rep-lication. However, emergent WNV can evade IFIT2 restrictionthrough 2�-O modification of the 5� non-translated region ofthe viral RNAmediated by themethyltransferase activity of theviral NS5 protein (23). These observations define IFIT2 as acritical host factor of IFN action and WNV restriction, andunderscore the IFIT2/WNV interaction as a critical virus/hostinterface governing innate antiviral immunity and infectionoutcome.IFIT2 is expressed after virus infection directly upon IRF-3

activation as well as upon IFN signaling because of the presenceof both IRF-3 and ISGF3 binding sites in the Ifit2 promoter (24,29–31). However, STAT1�/� mice failed to induce IFIT2expression in the CNS following Lymphocytic Choriomeningi-tis virus (LCMV) andWNV-infection in vivo (32). Importantly,IFN-induced expression of murine IFIT2 is dependent uponSTAT1 Ser-708 phosphorylation, as recently described bytenOever et al. (21). In this respect, Ifit2 is among a set of ISGs,including Adar1,Mx1, and Oas1b, and others, whose promot-ers lack a purine-rich region upstream of their ISRE, whichotherwise serves as STAT2 binding sites, wherein STAT1 Ser-708 phosphorylation is thought to increase the affinity forSTAT1 binding sites to confer gene expression by ISGF3 in theabsence of the STAT2 binding site (21). Notably, STAT1 Ser-708 phosphorylation is induced after IFN treatment of cells.However, the temporal relationship of Ser-708 phosphoryla-tion to other STAT1 phosphorylation sites during IFN-stimu-lation, innate immune responses, or WNV infection is notknown.We therefore conducted the current study to define thecell signaling pathway and STAT1 phosphorylation interac-tions that drive IFN-induced IFIT2 expression for the restric-tion of WNV infection. Our results reveal an IKK�-dependentpathway of STAT1 Ser-708 phosphorylation whose activationrequires IFN signaling for ISG expression and that plays a keyrole in the temporal regulation of STAT1 phosphorylation andthe expression of IFIT2 crucial to the control ofWNV infectionand immunity.

EXPERIMENTAL PROCEDURES

Cell Culture, Interferons, Viruses—Parental WT 2fTGHfibrosarcoma cells, U3A (2fTGH-derived mutant cells, defi-cient in STAT1) and U5A (IFRNAR-deficient, provided by Dr.

George Stark, Cleveland Clinic), immortalized humanPH5CH8 hepatocytes (provided by Dr. Nobuyuki Kato,OkayamaUniversity, Japan), BabyHamsterKidney (BHK) cells,and HEK293 cells were grown in DMEM supplemented with10% FBS, 2mML-glutamine, 1mM sodiumpyruvate, antibiotic-antimycotic solution, and 1� nonessential amino acids (com-plete DMEM). Primary mouse embryonic fibroblasts (MEFs)were isolated from IKK��/�, IRF-3�/�, IFNAR�/�, IFIT2�/�,and age-matched WT control mice as described previously (7,33), and grown inDMEM.Human IFN�-2a, human IFN-�, andmurine IFN� (PBL InterferonSource) were used at a concentra-tion of 100 IU/ml, whereas human IFN-� and IFN-�1 (PBLInterferonSource) were used at concentrations of 50 ng/ml and100 ng/ml, respectively. The Madagascar-AnMg798 strain ofWNV (WNV-MAD) was obtained from the World ReferenceCenter of Emerging Viruses and Arboviruses and passaged inVero cells as described previously (8). The Cantell strain ofSendai virus (SenV) was obtained from the Charles River Lab-oratory.Where indicated, cells weremock-treated, treatedwithIFN, or infected with either WNV-MAD at an m.o.i. of 1 orSenVat 100HAunits/ml for the indicated times before harvest-ing for immunoblot assay (8, 34). Cycloheximide (CHX, Sigma)and leptomycin B (LMB, Sigma) was used at a concentration of50 �g/ml and 100 nM, respectively. Pervanadate was preparedby mixing equal volumes of 50 mM H2O2 and 50 mM sodiumorthovanadate to make 50 mM pervanadate before adding itinto growth media for a final concentration of 50 �M.Transfection and Promoter-Luciferase Analyses—pFLAG-

IKK�, pFLAG-STAT1 WT, pFLAG-STAT1 Y701F, andpFLAG-STAT1 S727A expression plasmids were gifts fromDr.Curt Horvath (Northwestern University). pFLAG-STAT1mutant constructswere generated by site-directedmutagenesisusing the QuikChange XL site-directed mutagenesis kit (Strat-agene) and the following primers: 5�-CAAGACTGAGTTGAT-TGCTGTGTCTGAAGTTC-3� (forward) and 5�-GAACTTCA-GACACAGCAATCAACTCAGTCTTG-3� (reverse) for S708A;5�-CAAGACTGAGTTGATTGATGTGTCTGAAGTTCA-CCCTTCTAGAC-3� (forward) and 5�-GTCTAGAAGGGT-GAACTTCAGACACATCAATCAACTCAGTCTTG-3� (re-verse) for S708D; and 5�-GATGGCCCTAAAGGAACTGGA-GAGATCAAGACTGAGTTG-3� (forward) and 5�-CAACTC-AGTCTTGATCTCTCCAGTTCCTTTAGGGCCATC-3� (re-verse) for Y701E.pIFIT2-Lucwas a gift fromKineta, Inc. (Seattle). pISG15-Luc

and pIFN-�-Luc were described previously (34). pADAR1-CAT was a gift fromDr. Charles Samuel (University of Califor-nia-SanDiego), and pADAR1-Lucwas generated by subcloninginto the pGL3 Luciferase reporter vector. Transfections werecarried out for 16 h using FuGENE 6 transfection reagent(Roche) and following the manufacturer’s instructions. Forluciferase analyses, cells were cotransfected with either theIFN-�-Luciferase, ISG15-Luciferase, or ADAR1-Luciferaseconstruct; CMV-Renilla; and the indicated cDNA expressionplasmids. Cell extracts were collected 24 h post-infection orpost-treatment and analyzed for dual luciferase activity(Promega).In Vivo Mouse Infection—C57BL/6 (Bl6) WT and IKK��/�

mice were purchased from The Jackson Laboratory (21). Mice

IKK� Restriction of WNV Infection

DECEMBER 30, 2011 • VOLUME 286 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 44413

were bred in the animal facility at theUniversity ofWashingtonunder specific pathogen-free (SPF) conditions. Experimentsusing these animals were completed within the approval andguidelines by the University of Washington Institutional Ani-mal Care and Use Committee. IKK��/� and age-matched WTcontrol mice were inoculated subcutaneously in the left rearfootpad with 103 or 104 pfu of WNV-MAD as described previ-ously (12). Mice were monitored daily for morbidity and mor-tality. Clinical symptoms were numerically scored: 1, ruffledfur/lethargic/hunched/no paresis; 2, very mild to mild paresis;3, frank paresis in at least one hind limb or mild paresis in twohind limbs; 4, severe paresis, still retains feeling and possiblylimbic; 5, true paralysis; 6, moribund; 7, dead (12). For the invivo viral burden analysis, infectedmice were bled and perfusedwith 20 ml of PBS following euthanasia. Spleens were collectedand homogenized for immunoblot analysis as described previ-ously (12).Immunoblot Analysis—Protein extracts were prepared by

lysing cells in radioimmune precipitation assay buffer (50 mM

TrisHCl (pH7.5), 150mMNaCl, 0.5% sodiumdeoxycholate, 1%Nonidet P-40, 1 mM EDTA, and 0.1% sodium dodecyl sulfate)supplementedwith 1�Mokadaic acid, 1�Mphosphatase inhib-itor mixture II (Calbiochem), and 10 �M protease inhibitor(Sigma), followed by 4 °C centrifugation at 16,000 � g for 10min to clarify the lysate. Equivalent protein amounts were ana-lyzed by SDS-polyacrylamide gel electrophoresis followed byimmunoblotting. Affinity-purified rabbit polyclonal anti-STAT1 Ser-708 antibody was generated by repeat-immuniza-tion with the STAT1 Ser-708 phospho-specific peptide(YIKTELI{pS}VSEVHP, amino acids 701–714) (GenScript).The following primary antibodies were used for immunoblotanalyses: �-ADAR1 (Abnova); �-IRF-3 (M. David, UCSD);�-IFIT1, �-murine IFIT2 and �-murine IFIT3 (G. Sen, Cleve-land Clinic); �-ISG15 (A. Haas, Louisiana State University);�-WNV (Centers for Disease Control and Prevention); �-p-STAT1 Tyr-701, �-p-STAT1 Ser-727, �-STAT1, �-p-IRF-3(Cell Signaling Technology, Inc.); �-murine IRF-3 (Invitrogen);�-IKK� (Imgenex); �-PKR (Santa Cruz Biotechnology, Inc.);�-SenV (Biodesign International); �-FLAG (M2), and �-Tubu-lin (Sigma). HRP-conjugated goat anti-rabbit, goat anti-mouse,and donkey anti-goat (Jackson ImmunoResearch Laboratories,Inc.) were used as secondary antibodies.Immunoprecipitation—Following FLAG-STAT1 reconstitu-

tion of U3A cells and IFN-� treatment, cell extracts wereimmunoprecipitated using �-FLAG (M2)-conjugated agarosebeads (Sigma) for 2 h at 4 °C. Samples were then washed threetimes with radioimmune precipitation assay buffer before elu-tion. The eluate was heated for 5 min and analyzed by SDS gelelectrophoresis and immunoblot analysis. Subsequently, 20 �lof 50% slurry of protein G-agarose beads (Calbiochem) wasadded and incubated for 2 h at 4 °C. Beads were washed andeluted as described above. Clean-Blot HRP-conjugated immu-noprecipitation detection reagent (ThermoScientific) was usedas a secondary antibody for the immunoblot assay.

RESULTS

IKK� and IFIT2 Impose Restriction of WNV Infection—Todetermine the role of IKK� in IFIT2 expression during WNV

infection, we evaluated the IFIT2 abundance and virus replica-tion in primary mouse embryonic fibroblasts (MEF) isolatedfrom WT or IKK��/� mice. For these studies we utilized theavirulent lineage 2 Madagascar strain of WNV (WNV-MAD).As opposed to virulent lineage 1 strains, WNV-MAD lacks theability to block IFN signaling and is highly sensitive to theinnate immune antiviral actions of IFN (8), thus allowing stud-ies of IFN actions against WNV without confounding influ-ences of viral antagonism of the IFN response. As shown in Fig.1A, WNV infection induced the expression and accumulationof IFIT2 in amanner dependent on IKK�, but IFIT1, IFIT3, andPKR were induced by WNV regardless of IKK�. FollowingWNV infection of wild-type MEFs, IFIT2 was expressed 48 hpost-infection, whereas IFIT1 and IFIT3 expression was firstobserved by 24 h (supplemental Fig. S1). However, virus-in-duced expression of IFIT2 was severely attenuated in IKK��/�

MEF infected with WNV, whereas induction of IFIT1 andIFIT3 expression remained comparable withWT controls (Fig.1A). Furthermore, in absence of IKK�, we observed delayed andimpaired IFIT2 expression following IFN-� stimulation, dem-onstrating that IKK� was important for IFIT2 induction inMEFs. In contrast, IFN-� stimulation efficiently induced theexpression of IFIT1, IFIT3, and PKR regardless of IKK� expres-sion (Fig. 1B). To further assess the specific role of IKK� inregulating ISG expression, we evaluated the ISG promoterinduction inHEK293 cells ectopically expressing IKK�. Ectopicoverexpression of IKK� has been shown to induce its multim-erization through the coiled-coil domain, causing its trans-au-toactivation. Trans-autoactivation of IKK� results in its signal-ing to activate downstream substrates (such as IRF-3, I�B�, andAKT) and induce the expression of target genes (35–37). Wefound that ectopic expression of IKK� induced activation ofIFN-�-promoter (which contains IRF-3 but not the ISGF3binding site) occurred in a dose-dependent manner, whereastreatment of cells with exogenous IFN-� did not induce pro-moter expression, as expected (Fig. 1C, p � 0.0025, top leftpanel). Importantly, we observed a dose-dependent promoteractivation of ADAR1 and IFIT2 (Fig. 1C, p � 0.025, bottompanels) but not ISG15 upon IKK� ectopic expression (C, p �0.025, top right panel), further demonstrating the role of IKK�in induction of an ISG subset recently shown to be sensitive toIKK� and STAT1 Ser-708 phosphorylation (21). UnlikeADAR1, whose promoter contains ISRE but not IRF-3 bindingsites, the IFIT2 promoter contains both sites, each of which areIKK� responsive. Our results show that IKK� is essential forboth virus-induced and IFN-induced IFIT2 expression, dem-onstrating dual roles of IKK� to induce innate effector ISGexpression.Because IFIT2 expression is induced by WNV infection, we

assessed the antiviral actions of IFIT2 in restricting WNV-MAD infection inMEFs fromWT and IFIT2�/� mice. Culturesupernatants were collected 0, 6, 24, and 48 h post-infection,where time 0 represents the input virus. Analysis of single-stepvirus growth revealed that WNV-MAD replication wasenhanced in cells lacking IFIT2 compared with WT cells, thusvalidating IFIT2 function as aWNV restriction factor (Fig. 1D)(23). These results define IFIT2 as an IKK�-dependent ISG

IKK� Restriction of WNV Infection

44414 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 52 • DECEMBER 30, 2011

whose expression is induced by WNV infection and IFN-� torestrict WNV growth in primary cells.Virus Infection Induces Delayed STAT1 Ser-708 Phos-

phorylation—Because IFIT2 is induced by WNV infection in amanner dependent on IKK�, we sought to characterize theIKK� and STAT1 Ser-708 phosphorylation kinetics in responseto cell treatment with IFN-�. IKK� has been shown previouslyto directly phosphorylate STAT1 Ser-708 in vitro (21). Wetherefore generated novel phospho-specific, affinity-purifiedpolyclonal antibody against a phosphorylated peptide repre-senting phospho-STAT1 Ser-708 (�-p-STAT1 Ser-708, sup-plemental Fig. S2, A–C). We used this antibody to assessSTAT1 Ser-708 phosphorylation status after IFN-�-treatment,confirming that IKK��/� MEFs were deficient in IFN-�-in-duced Ser-708 STAT1 phosphorylation. In contrast, IFN-�-in-duced STAT1 Tyr-701 and Ser-727 phosphorylation occurredindependently of IKK� (Fig. 2A). IKK�-mediated STAT1 Ser-708 phosphorylation is independent of its role in IRF-3 activa-tion, as activated IRF-3 efficiently translocated into the nucleus

following SenV infection in the absence of IKK� (supplementalFig. S3). These results demonstrate the specificity of the �-p-STAT1 Ser-708 antibody and confirm that IFN-inducedSTAT1 Ser-708 phosphorylation is dependent on IKK�expression.To further determine the kinetics of STAT1 Ser-708 phos-

phorylation in response to RNA virus infection, we analyzedp-STAT1 Ser-708 abundance during infection of HEK293 cellsby the prototypic Paramyxovirus, SenV, orWNV-MAD (Fig. 2,B and C). Immunoblot analysis revealed that phosphorylatedSTAT1 (p-STAT1) Ser-708 occurred at later time points duringthe infection cycle of either virus, happening between 18 to 24 hpost-infection with SenV and 72 h post-infection with WNV-MAD. In both SenV andWNV infection models, phosphoryla-tion of Ser-708 appeared to be occurring much later than thecanonical phosphorylation of STAT1 Tyr-701, which began at6 h post-SenV infection and as early as 36 h post-WNV infec-tion. Moreover, phosphorylation of STAT Tyr-701 was pre-ceded by detectable levels of phosphorylated/activated p-(IRF-

FIGURE 1. IKK� and IFIT2 impose restriction on WNV infection. WT or IKK��/� MEFs were mock-infected or infected with WNV-MAD at an m.o.i. of 1 (A) andmock-stimulated or stimulated with 100 IU/ml IFN-� (B). Protein lysate was collected at the indicated times and immunoblotted using IFIT2, IFIT3, IFIT1, PKR,and IKK� antibodies. Tubulin was used as loading control. C, HEK293 cells were cotransfected with pCMV-Renilla and either pIFN-�-Luciferase (top left panel),pISG15-Luciferase (top right panel), pADAR1-Luciferase (bottom left panel), or pIFIT2-Luciferase (bottom right panel). 16 h later, cells were either mock-stimu-lated (vector cotransfection); stimulated by transfection with 25 ng, 50 ng, or 100 ng of an IKK� expression plasmid; or treated with 100 IU/ml IFN-�. Cells wereharvested 48 h post-transfection, and luciferase expression was measured and normalized to Renilla. The relative luciferase value was calculated as foldinduction over induction of the vector that was set to 1. Statistical analysis was performed with Student’s t test. D, WT or IFIT2�/� MEFs were infected withWNV-MAD at an m.o.i. of 5. At the indicated time points post-infection, culture supernatants were collected, and virus titers were determined by plaque assayon BHK cells. C and D, data are mean � S.D. of three independent experiments performed in triplicate. p values were calculated using Student’s t test todetermine statistical significance.

IKK� Restriction of WNV Infection

DECEMBER 30, 2011 • VOLUME 286 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 44415

3), which is consistent with endogenous IFN being expressed todrive IFNAR signaling of STAT1 phosphorylation (Fig. 2B).Together, these results demonstrate that both SenV andWNV-MAD infections stimulate STAT1 Ser-708 phosphorylationlate in the virus replication cycle and that that IFIT2 expressionassociates with IFN-induced accumulation of p-STAT1Ser-708.Type I, Type II, and Type III IFNs Induce STAT1 Ser-708

Phosphorylation—To determine whether different classes ofIFNs stimulate STAT1 Ser-708 phosphorylation, we com-pared the ability of type I, II, and III IFNs to induce Ser-708phosphorylation in 2fTGH human fibrosarcoma cells aftertreatment with IFN-� (type I IFN), IFN-� (type II IFN), orIFN-� (type III IFN). We found that IFN-� and IFN-� treat-ment of cells induced STAT1 S708 phosphorylation after theonset of Tyr-701 and Ser-727 phosphorylation at a late timepost-treatment and similar to the kinetics of p-STAT1 Ser-708 accumulation duringWNV and SenV infection (Fig. 3, Aand B). In IFN-�-treated cells, STAT1 was immediatelyphosphorylated at Tyr-701 and Ser-727 within 10 and 30minafter treatment, respectively. In contrast, STAT1 Ser-708phosphorylation was first detectable at 16 h and peaked at24 h post-treatment, which coincides with diminishing levelof Tyr-701 phosphorylation and expression of ADAR1 (likeIFIT2, an IKK�-dependent ISG). Likewise, when treated with

IFN-�, 2fTGH cells induced STAT1 Tyr-701 and Ser-727phosphorylation within 10 min and through 16 h after treat-ment, whereas STAT1 Ser-708 phosphorylation occurred at16 h post-treatment. To evaluate the ability of type III IFN tostimulate STAT1 Ser-708 phosphorylation, we examined thephosphorylation status of Ser-708 in lysates of PH5CH8cells, an immortalized hepatocyte cell line that expressesendogenously the type III IFN receptor. Interestingly, wefound that PH5CH8 cells treated with IFN-�1 exhibitedfaster STAT1 Ser-708 phosphorylation kinetics comparedwith IFN-� treatment (Fig. 3C, lanes 5–7 and 10), withp-STAT1 Ser-708 accumulating within 2 h after IFN-�1treatment. In addition, we observed that p-STAT1 Ser-727was weakly phosphorylated following IFN-�1 treatmentcompared with IFN-�. Furthermore, STAT1 Tyr-701 phos-phorylation was no longer sustained after 6 h of IFN-�1treatment, a finding which contrasted with the phosphory-lation kinetics observed in IFN-�-treated cells. In all cases,we found that IFN treatment induces phosphorylation ofSTAT1 at the Tyr-701 residue before phosphorylation isdetected on Ser-708 (Fig. 3, A–C). Together, these observa-tions demonstrate that type I, II, and III IFN are all able toinduce STAT1 phosphorylation at residue Ser-708, albeitwith different kinetics.

FIGURE 2. Virus infection induces delayed STAT1 Ser-708 phosphorylation. A, WT or IKK��/� MEFs were mock-stimulated or stimulated with 100IU/mlIFN-�. Protein lysate was collected 16 h post-IFN stimulation and immunoblotted using p-STAT1 Ser-708, p-STAT1 Ser-727, p-STAT1 Tyr-701, and total STAT1antibodies. B and C, Sendai and WNV virus infections induce STAT1 Ser-708 phosphorylation. HEK293 cells were infected with 100 HA units/ml SenV (B) or WestNile virus strain Madagascar (WNV-MAD) (C) at an m.o.i. of 1. At the indicated times following infection, protein lysates were collected and immunoblotted forp-STAT1 Ser-708, p-STAT1 Tyr-701, total STAT1, p-IRF-3, total IRF-3, IFIT1, and SenV or WNV. Tubulin and GAPDH were used as loading controls.

IKK� Restriction of WNV Infection

44416 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 52 • DECEMBER 30, 2011

Signaling through IFNAR Is Required for STAT1 Ser-708Phosphorylation Following Type I IFN Treatment or VirusInfection—To evaluate the signaling requirements for WNV-MAD-inducedSTAT1Ser-708phosphorylation,we infectedWT,IRF-3�/�, and IFNAR�/� MEFs with WNV-MAD (m.o.i. � 1)and analyzed STAT1 tyrosine and serine phosphorylation byimmunoblot assay. As expected, phosphorylation of STAT1 atTyr-701 and Ser-727 were ablated in the absence of IFNAR.Phosphorylation at Ser-708 was similarly blocked despite thehigh abundance of phospho/active IRF-3 and IFN-� secretion

of IFNAR�/� MEFs, indicating that active signaling throughIFNAR is also required for STAT1 Ser-708 phosphorylationduringWNV-MAD infection (Fig. 4A, lanes 9–12, supplemen-tal Fig. S4). Similarly,WNV-infectedU5Ahuman fibrosarcomacells, which lack IFNAR2, also failed to induce STAT1 Ser-708phosphorylation compared with parental 2fTGH cells (Fig. 4B)(38). Next, we evaluate the IRF-3-signaling requirement forWNV-MAD-induced STAT1Ser-708 phosphorylation. In IRF-3�/� MEFs, there was a lack of STAT1 Ser-708 phosphoryla-tion during WNV-MAD infection (Fig. 4A, lanes 5–8), a find-ing that might be explained by the cellular defect in IFN-�production in the absence of IRF-3 (supplemental Fig. S4) (39).Thus, to assess the possible outcome because of loss of IFN-�production and its impact on p-STAT1 Ser-708 accumulationin these cells, we compared p-STAT1 Ser-708 abundance inWT, IRF-3�/�, and IFNAR�/� MEFs after treatment with 100IU/ml IFN-�. p-STAT1 Ser-708 levels in IRF-3�/� MEFs weresimilar or greater to the level found in WT MEFs after IFN-�treatment (Fig. 4C, lanes 3–4). However, IFN-� failed to induceSTAT1 phosphorylation in IFNAR�/� MEFs and U5A cells(Fig. 4C, lanes 5–6, and Fig. 4D). Thus, IFNAR signalingbut not IRF-3 signaling is required for STAT1 Ser-708phosphorylation.STAT1 Ser-708 Phosphorylation Requires de Novo Protein

Synthesis—Because STAT1 Ser-708 phosphorylation firstrequires IFN signaling, we sought to determine whether anIFN-responsive factor(s) might be required to induce p-STAT1Ser-708 accumulation and the subsequent expression of IFIT2.To test this notion, we assessed p-STAT1Ser-708 abundance in2fTGHcells thatwere eithermock-treated or treatedwithCHXfor 30 min prior to a 16-h IFN-� treatment time course. Weobserved that IFN-�-induced STAT1 phosphorylation at Ser-708, but not Tyr-701, is abrogated when de novo protein syn-thesis is blocked (Fig. 5A, lanes 6–10, supplemental Fig. S5A,lane 3).We found thatwhen cells were pretreatedwithCHX for16 h and then subsequently treated with IFN-� for 1 h,p-STAT1 Tyr-701 still accumulated to high levels, revealingthat available STAT1 molecules can be readily phosphorylatedat Tyr-701 during long-term protein synthesis inhibition, andthat IFN receptor signaling remains intact under these condi-tions. Thus, STAT1 phosphorylation on Tyr-701 does notrequire de novo protein synthesis (Fig. 5A, lane 11). Theseobservations also demonstrate that the CHX-treated cells wereviable and responsive following 16 hours of protein synthesisinhibition (supplemental Fig. S5B). Similar to treatment withIFN-�, IFN-�-induced accumulation of p-STAT1 Ser-708, butneither Tyr-701 nor Ser-727 phosphorylation, was also blockedin the presence of CHX (data not shown). Taken together, thesedata show that de novo protein synthesis is required for STAT1Ser-708 phosphorylation but notTyr-701 or Ser-727 phosphor-ylation following treatment of cells with IFN-� or IFN-�. Thus,STAT1 Ser-708 phosphorylation is induced and regulatedthough the actions of ISG product(s) whose expression pre-cedes p-STAT1 Ser-708 accumulation. Furthermore, we foundthat STAT1Ser-708 phosphorylation only occurredwhenCHXwas added more than 9 h after the addition of IFN-� (Fig. 5A,lanes 7–9), suggesting that the ISG product(s) required forSTAT1 Ser-708 phosphorylation is synthesized between 9 and

FIGURE 3. Type I, type II, and type III IFNs induce STAT1 Ser-708 phosphor-ylation. 2fTGH cells were mock-treated or treated with 100 IU/ml IFN-� (A) or50 ng/ml IFN-� (B). C, PH5CH8 cells were mock-treated (lane 1), treated with100 ng/ml IFN-�1 (lanes 2-6), or 100 IU/ml IFN-� (lanes 7-9). Protein lysate wascollected at respective time points following IFN treatment and immuno-blotted for p-STAT1 Ser-708, p-STAT1 Tyr-701, p-STAT1 Ser-727, total STAT1,ADAR1, and IFIT1.

IKK� Restriction of WNV Infection

DECEMBER 30, 2011 • VOLUME 286 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 44417

12 hours after the initiation of IFN-� treatment. In contrast,IFN-� stimulation of STAT1 Tyr-701 phosphorylationoccurred regardless of CHX treatment, indicating that therequirement for a de novo synthesized IFN-responsive geneproduct is specific to Ser-708 phosphorylation (Fig. 5A, lanes6-11). Because STAT1 itself is an ISG, we assessed whether ornot de novo STAT1 expression is required for Ser-708 phos-phorylation. We found that ectopic overexpression of STAT1does not induce its phosphorylation at Ser-708. Moreover, inthe presence of IFN-�, we did not observe an acceleration ofSTAT1 Ser-708 phosphorylation kinetics when compared withvector-transfected control cells (data not shown), demonstrat-ing that de novo STAT1 expression does not immediately resultin Ser-708 phosphorylation. Thus, an IFN-responsive factor(s)but not STAT1 itself is the primary ISG product(s) drivingSTAT1 Ser-708 phosphorylation and the IFN-induced expres-sion IFIT2.STAT1 Ser-708 Phosphorylation Requires STAT1 Tyrosine

Dephosphorylation and Nuclear Export—Given the differentkinetics of STAT1 phosphorylation at various phospho-resi-

dues following IFN-� treatment and the requirement for ISGexpression to drive p-STAT1 Ser-708 accumulation, we inves-tigated whether the phosphorylation of STAT1 Tyr-701 andSer-708 are linked. We assessed the impact of STAT1 Tyr-701phosphorylation on the accumulation of p-STAT1 Ser-708 bypretreatment of cells to sustain STAT1 Tyr-701 phosphoryla-tion upon subsequent treatment with IFN-�. In the absence ofinhibitor, IFN-� stimulation resulted in early induction ofp-STAT1 Tyr-701. However, p-STAT1 Tyr-701 levels dimin-ished after 16 h of IFN-� stimulation despite increased totalSTAT1 abundance (Figs. 5B, lanes 1-3, and 3A). Pretreatmentof 2fTGH cells with CRM1 nuclear export inhibitor LMB orprotein tyrosine-phosphatase (PTP) inhibitor pervanadate for1 h before the start of IFN treatment resulted in the sustainedaccumulation of p-STAT1 Tyr-701 within IFN-treated cells(Fig. 5B, lanes 4-9, and Refs. 40–42). In agreement with a pre-vious report, pervanadate treatment of cells induced low levelSTAT1 activation in the absence of IFN-stimulation and fur-ther enhanced STAT1 tyrosine phosphorylation following IFNstimulation (Fig. 5B, lanes 7-9, and Ref. 42). Importantly, there

FIGURE 4. Signaling through IFNAR is required for STAT1 Ser-708 phosphorylation following type I IFN treatment or virus infection. WT, IRF-3�/�, orIFNAR�/� (A) and parental 2fTGH cells or their derivative U5A cells (which lack IFNAR) (B) were infected with WNV-MAD at an m.o.i. of 1. Protein lysates werecollected at the indicated time points and immunoblotted for p-STAT1 Ser-708, p-STAT1 Tyr-701, p-STAT1 Ser-727, total STAT1, p-IRF-3, total IRF-3, and WNV.C and D, the same cells were also mock-stimulated or stimulated with 100 IU/ml IFN-� for 6 or 16 h. Protein lysates were collected at the indicated time pointsand immunoblotted for p-STAT1 Ser-708, p-STAT1 Tyr-701, p-STAT1 Ser-727, total STAT1, p-IRF-3, total IRF-3, IFIT2, IFIT3, and IFIT1. Asterisk, nonspecific band.

IKK� Restriction of WNV Infection

44418 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 52 • DECEMBER 30, 2011

was an absence of p-STAT1 Ser-708 concomitant with the lackof IFN-induced ADAR1 expression (like IFIT2, an IKK�- andp-STAT1 Ser-708-dependent ISG (21)) in cells pretreated witheither LMB (Fig. 5B, lanes 4–6) or pervanadate (Fig. 5B, lanes7-9). However, STAT1 Ser-727 phosphorylation and non-IKK�-dependent ISG expression were effectively induced uponIFN-� treatment of these cells. This observation suggests thatSTAT1 phosphorylation at the Tyr-701 and Ser-708 residuesare mutually exclusive and that removal of Y701 phosphoryla-tion and subsequent STAT1 nuclear export are prerequisitesthe phosphorylation of STAT1 on Ser-708.To further assess the relationship of p-STAT1 Tyr-701 and

Ser-708, we evaluated STAT1 site-specific phosphorylation inSTAT1-negative U3A cells reconstituted with transfectedFLAG-tagged constructs containing eitherWT FLAG-STAT1,FLAG-STAT1 Y701E phosphomimetic, FLAG-Y701F phos-phomutant, FLAG-S708A phosphomutant, FLAG-S708Dphosphomimetic, FLAG-S727A phosphomutant, or a FLAGvector control. We assessed STAT1 phosphorylation at eachsite after cells were treated with IFN-� for 16 h. We found thatalthough STAT1 Tyr-701 phosphorylation was absent in IFN-treated cells reconstituted with FLAG-STAT1 S708D, it waspresent in cells reconstituted with STAT1 S708A (supplemen-

tal Fig. S6, lane 6). Furthermore, we detected STAT1 Ser-708phosphorylation only in cells reconstituted with the FLAG-STAT1 Y701F or FLAG-STAT1 S708D, the latter observationdefining the FLAG-STAT1 S708D construct as a direct phos-phomimetic recognized by our anti-phospho STAT1 Ser-708antibody. STAT1 Ser-708 phosphorylation was not detected incells expressing FLAG-STAT1 WT or FLAG-STAT1 S727A,both of which were phosphorylated on Tyr-701 (supplementalFig. S6, lanes 2 and 7). In fact, we found that each of theseconstructs becomes immediately phosphorylated at Tyr-701upon IFN treatment and are sustained as such throughout thecourse of IFN stimulation (data not shown). Cells reconstitutedwith FLAG-STAT1 Y701E failed to display Ser-708 phosphor-ylation (supplemental Fig. S6, lane 3). Thus, p-STAT1 Ser-708likely takes place only after Tyr-701 dephosphorylation andnuclear export, which occurs approximately at 16 h post IFN-�stimulation.IKK�Mediates IFIT2Expression andProtection againstWNV

Pathogenesis in Vivo—To determine the role of IKK� andSTAT1 Ser-708 phosphorylation in IFIT2 expression and pro-tection against WNV infection in vivo, we examined theresponse ofWTand IKK��/�mice toWNVchallenge.WTandIKK��/� mice were challenged with 103 pfu WNV-MAD bysubcutaneous injection into the footpad. Clinical symptomswere monitored daily during the course of infection to observethe occurrence of disease and neurovirulence (12).When com-pared with the WT controls, IKK��/� mice displayed earlierneurological symptoms, a higher degree of neurovirulence, anda failure to recover from acuteWNV-MAD infection (Fig. 6A).Furthermore, we observed lack of sustained IFIT2 expression inthe spleen of IKK��/� mice during infection, and this associ-ated with earlier virus entry to the spleen compared with WTcontrols (Fig. 6B). For comparison, we also challengedWT andIKK��/� mice with the virulent/emergent lineage 1 Texas 02strain ofWNV (WNV-TX). This strainmediates a robust blockto IFN signaling while evading the antiviral actions of IFIT2 (8,23). As expected, we observed a similar but more rapid pathol-ogy defined by neurovirulence among WT and IKK��/� miceinfected with WNV-TX (data not shown). These observationsreveal a dependence of IKK� for IFIT2 expression duringWNVinfection in vivo and demonstrate that IKK� plays a role in pro-gramming the innate immune response for the expression ofIFIT2 and the control WNV infection. Our data also under-score the pathogenic outcome ofWNV infection linked to viralevasion of IFN defenses.

DISCUSSION

Our study identifies IFIT2 as an innate immune effector genethat can restrict WNV replication and defines the IKK�-medi-ated signaling pathway of IFN action that drives the expressionof IFIT2 and a subset of ISGs through phosphorylation ofSTAT1Ser-708. Furthermore,we reveal that this IKK�pathwayis dependent on an ISG product(s) to stimulate the IKK�-di-rected STAT1 Ser-708 phosphorylation at late times in the IFNresponse. Recent studies have demonstrated the importance ofa variety of ISGs in controllingWNV infection, such as Viperin,IFITM2, IFITM3, ISG20, PKR, and IFIT2 (23, 43). Moreover,the IFIT family members have been shown to suppress protein

FIGURE 5. STAT1 Ser-708 phosphorylation requires de novo protein syn-thesis, STAT1 tyrosine dephosphorylation, and nuclear export. A, 2fTGHcells were mock-treated (-CHX, lanes 1-5) or treated with CHX (�CHX, lanes6-11) to block protein synthesis. At 30 min (lanes 6-10) or 16 h (lane 11) follow-ing CHX treatment, cells were mock-stimulated (M) or stimulated with IFN-�.Cells were harvested at 10 min as well as 1, 6, and 16 h post-IFN stimulationand immunoblotted to detect p-STAT1 Ser-708, p-STAT1 Tyr-701, total STAT1,ISG15, and IFIT1. B, 2fTGH cells were not treated (NT, lanes 1-3), pretreatedwith 100 nM LMB (lanes 4-6), or 50 mM pervanadate (Van, lanes 7-9) 1 h beforemock stimulation or stimulation with 100 IU/ml IFN-�. Cells were harvested at1 and 16 h post-stimulation. Immunoblot analysis was performed usingp-STAT1 Ser-708, p-STAT1 Tyr-701, p-STAT1 Ser-727, total STAT1, ADAR1, andIFIT1 antibodies.

IKK� Restriction of WNV Infection

DECEMBER 30, 2011 • VOLUME 286 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 44419

synthesis, thus restricting replication of Alphavirus, Papillo-mavirus, and hepatitis C virus (28, 44, 45). Our results nowshow that IFIT2 can restrict WNV growth in vitro and demon-strate that its expression within an IKK�-dependent innateimmune effector pathway associates with the control of virusspread and pathogenesis in vivo. Although themagnitude of theincrease of WNV-MAD replication in IFIT2�/� cells was only10-fold (one log), it is notable that this difference was statisti-cally significant and caused by loss of a single ISG of severalhundred known ISGs, many of which might restrict WNVinfection (43). These observations indicate the importance ofIFIT2 in controlling the growth ofWNVand indeedmay in partexplain the immune protection and lack of pathogencity afterinfection by low virulence WNV strains such as WNV-MADand others in animals (8, 46, 47).IKK� has multiple roles in activating the innate immune

response to virus infection, including the phosphorylation andactivation of IRF-3, which leads to IFN-� production and phos-phorylation of STAT1 at residue Ser-708 following IFNAR sig-

naling (21, 48, 49). The role of IKK� in STAT1 phosphorylationis independent of its role in IRF-3 activation, a role which isredundant with the related kinase TBK1. MEFs lacking TBK1are deficient in IRF-3 phosphorylation, indicating that TBK1and not IKK� is the dominant kinase for IRF-3 activation, atleast in MEFs (50). We conclude that IKK� functions to induceSTAT1 Ser-708 phosphorylation and a specific ISG expressionsignature that includes IFIT2 inWNV-infected cells. This con-clusion is supported by our findings that ectopic overexpres-sion of IKK� alone in human cells specifically stimulated IFIT2and ADAR1 promoter induction, whereas IFN-induced IFIT2expression is strictly linked to STAT1 Ser-708 phosphorylationand is dependent on IKK� in vitro and in vivo (see Figs. 1, 2, and6) (21). Ectopic overexpression of a kinase such as IKK� inducesits trans-autoactivation facilitated by its multimerization,therefore bypassing the requirement for upstream signaling(35, 36). In agreement with previous reports, IKK� overexpres-sion also induces IFN-� promoter activation (Fig. 1C and Ref.36), suggesting that general activation of IKK� target genesoccurs upon its overexpression. However, IKK� expressionalone did not induce the ISG15 promoter, an ISG that can beinduced through canonical ISGF3 function, which insteadrequired IFN treatment (see Fig. 1). These data suggest that theexpression of a specific ISG subset that includes IFIT2, ADAR1,and others directly depends on IKK�, thus supporting the novelrole for IKK� in antiviral immunity (21).

Our data now implicate this IKK�-dependent pathway andits specific expression of IFIT2 as important components of theinnate immune response to WNV infection and indicate thatthis response is governed by IKK� phosphorylation of STAT1Ser-708. IKK��/� MEFs, which are deficient in STAT1 Ser-708phosphorylation, showed defects in IFN-induced IFIT2 expres-sion but not in other related ISGs, including IFIT3 and IFIT1.We found that STAT1 Ser-708 phosphorylation was inducedby type I, II, and III IFNs in addition to being induced duringinfection by WNV or SenV. Moreover, the kinetics of IFN-induced STAT1 Ser-708 phosphorylation varied from phos-phorylation at Tyr-701 and Ser-727. Although STAT1 Tyr-701and Ser-727 phosphorylation occurred immediately followingtype I and type II IFN stimulation as known previously(reviewed in Ref. 15), Ser-708 phosphorylation occurred later,16 h post-IFN treatment. In comparison, type III IFN-in-duced phosphorylation of STAT1 Ser-708 occurred more rap-idly. This observation agrees with a previous report that dem-onstrated the differential kinetics and duration of JAK-STATsignaling activity induced by type I and III IFN (51) and suggeststhat antiviral immune actions of these IFNs are each mediatedin part through STAT1 Ser-708-responsive ISGs. Similarly,STAT1 Ser-708 phosphorylationwas induced at later times fol-lowing RNA virus infection, with delayed kinetics associatedwith WNV-MAD compared with SenV infection, and likelybecause of the slower growth rate and IFN induction of theformer. These observations indicate thatWNV and likely RNAvirus infections in general indirectly stimulate STAT1 Ser-708phosphorylation via viral induction of IFN production from theinfected cell, which then stimulates STAT1 Ser-708 phosphor-ylation through the actions of IKK�.

FIGURE 6. IKK� mediates IFIT2 expression and protection against WNVpathogenesis in vivo. WT Bl6 and IKK��/� mice were mock-infected (PBSonly) or infected with 103 pfu of WNV-MAD subcutaneously through footpadinjection. A, mice were monitored and scored daily for clinical symptoms over17 days. Clinical scores from four representative mice per group weregraphed. B, spleens from WT or IKK��/� mice, mock-infected or infected withWNV-MAD, were collected at days 4, 6, and 12 post-infection. Protein lysateswere extracted by homogenizing spleens with radioimmune precipitationassay buffer and immunoblotted using p-STAT1 Ser-708, p-STAT1 Tyr-701,p-STAT1 Ser-727, total STAT1, IFIT2, IFIT1, WNV, and IKK� antibodies. Theimmunoblot analysis panel is a representative from four mice per infectiongroup.

IKK� Restriction of WNV Infection

44420 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 52 • DECEMBER 30, 2011

We found that active signaling through the type I IFN recep-tor was required for IFN-�- and virus-induced STAT1 Ser-708phosphorylation. IFN-induced STAT1 Ser-708 phosphoryla-tion, however, did not require IRF-3 expression. These obser-vations are consistent with further data that IKK�-dependentIFIT2 induction can occur independently of IRF-3 (see Fig. 4C).However, during the course of virus infection, STAT1 Ser-708phosphorylation failed to take place in the absence of IRF-3because of a lack of IFN-� induction, secretion, and signaling.Indeed, de novo protein synthesis downstream of IFN signalingwas required for STAT1 Ser-708 phosphorylation, suggestingthat one or more ISG product signals IKK� to catalyze STAT1Ser-708 phosphorylation. This requirement of de novo IFN-induced factor synthesis for STAT1 Ser-708-responsive ISGexpression can be bypassed by IKK� overexpression, whichinduces its autoactivation, suggesting that an IKK� activatorISG would function upstream of IKK� (see Fig. 1C). Althoughwe have yet to determine the IFN-inducible factor that pro-motes Ser-708 phosphorylation, on the basis of CHX pulse-chase experiments, it appears to be synthesized between 9 and12h after IFN-� stimulation.More detailed time course-depen-dent transcriptome profiling experiments may narrow down alist of candidate ISGs that directly or indirectly activate thekinase activity of IKK� that is responsible for Ser-708phosphor-ylation. Potential candidates could include the IFN-inducedprotein kinase PKR, which interacts with STAT1 withoutdirectly phosphorylating the Tyr-701 residue (52) and restrictsWNV infection in cells and in vivo (53, 54), and p38, which hasbeen implicated in ISRE activation following type I IFN stimu-lation but is not required for IFN-dependent STAT1Tyr-701 orSer-727 phosphorylation (55–57). Alternatively, protein phos-phatases or non-enzymatic ISG products might be involved inmodulating IKK� action and STAT1 Ser-708 phosphorylationeither through regulation of a signaling network of IKK� con-trol or through direct binding to signaling factors of IKK� rele-vance or IKK�.

Our studies suggest that the order of STAT1 phosphoryla-tion during the course of IFN stimulation could be an importantcontributor to the kinetics of ISG expression as STAT1Tyr-701phosphorylation temporally precedes Ser-708 phosphorylationand the induction of IFIT2 expression. Moreover, we observedminimal Ser-708 phosphorylation in IFN-treated cells underconditions of pervanadate or LMB treatment, which blocksSTAT1 tyrosine dephosphorylation andnuclear export, respec-tively (see Fig. 5B). Additionally, these treatments result in sus-tained Tyr-701 phosphorylation of STAT1. These observationssuggest that Tyr-701 and Ser-708 could be mutually exclusiveon the same molecule. Consistent with this, STAT1 moleculesthat are phosphorylated at Tyr-701 following IFN-� treatmentare not phosphorylated on Ser-708 (supplemental Fig. S6).These observations suggest one of two possible scenarios ofSTAT1 phosphorylation kinetics in which 1) Tyr-701 and Ser-708 phosphorylation cannot occur simultaneously in a singleSTAT1 molecule, whereas Ser-727, which is more distantlylocated, can; or 2) STAT1molecules phosphorylated atTyr-701cannot dimerize with STAT1molecules phosphorylated at res-idue Ser-708. We favor the former hypothesis because of theclose proximity of the Tyr-701 and Ser-708 residues, as phos-

phorylation at Tyr-701 may result in steric hindrance of Ser-708 phosphorylation. Importantly, we note that Tyr-701 phos-phorylation has been shown to diminish at later time pointsafter IFN stimulation or virus infection because of nuclearSTAT1 acetylation and dephosphorylation of Tyr-701 by tyro-sine phosphatase TCP45, which has been reported previously(40, 58, 59). Additionally, chromatin-bound STAT1 can bephosphorylated at Ser-727, resulting in its sumoylation byUBC9 (60, 61). As unphosphorylated STAT1 cycles back to thecytoplasm via CRM1-mediated nuclear export, acetylation andsumoylation results in STAT1 latency by inhibiting IFN-in-duced STAT1 Tyr-701 phosphorylation, which should thenpermit Ser-708 phosphorylation (58, 59, 61). These studies con-cluded that non-tyrosine phosphorylated STAT1 is the“unphosphorylated STAT1,” which functions to sustainexpression of some ISGs (62, 63). Our findings now suggest thatthe actual nature of these unphosphorylated STAT1 entitiesmay be STAT1 phosphorylated at Ser-708, thus promoting theexpression of a specific subset of ISGs whose expression occurslater after IFN treatment, such as IFIT2, thus “sustaining” theIFN response. We therefore propose a model of early and latetype I IFN response programs (Fig. 7). Early after type I IFNstimulation, STAT1 is phosphorylated at Tyr-701, translocatesto the nucleus, and induces expression of IKK�-independentISGs. Following tyrosine dephosphorylation, STAT1moleculesare exported back to the cytoplasm. At a later time, as yet unde-termined IFN-inducible factor(s) activate the IKK�-mediatedSTAT1 phosphorylation at the Ser-708 residue, whichresults in sustained expression of IKK�-dependent ISGs,including IFIT2. Similar to the actions of IFIT2 againstWNVinfection, we propose that IKK�-dependent ISGs includegenes whose products direct antiviral and immune-modula-tory actions to mediate innate immunity. Defining thenature of these ISGs within the innate immune response toWNV infection will be an important contribution towardidentifying therapeutic targets for enhancement of immuneprotection against WNV and other flaviviruses. Therefore, agenomics-based assessment of the response to WNV infec-tion in WT and IKK��/� mice, as well as a targeted chroma-tin immunoprecipitation assay and analyses to definep-STAT1 Ser-708-responsive genes, is warranted for futurestudies aimed at characterizing this novel IKK�-dependentpathway of innate immunity.The importance of IFIT2 and the IKK� pathway of ISG

induction is underscored by the observation that virulentWNVeffectively suppresses IFIT2 function to ensure efficient virusreplication (23). Moreover, we have shown that a WNV strainthat specifically lacks the ability to modulate the effect of IFITgenes is attenuated in WT mice (23, 24, 26). These observa-tions, coupled with the present study showing that MEFs lack-ing IFIT2 support greater replication ofWNV-MAD, a strain ofWNV that only inefficiently antagonizes the antiviral effects ofIFN (8), define IFIT2 as a innate immune effector gene thatrestricts WNV replication. Our studies also examined the sig-nificance of STAT1 Ser-708 phosphorylation during the courseof infection in vivo, through infection of IKK��/� mice withWNV-MAD orWNV-TX, the latter being the emergent strainof WNV that is highly virulent and now circulates in North

IKK� Restriction of WNV Infection

DECEMBER 30, 2011 • VOLUME 286 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 44421

America (8). Although the WNV-MAD strain does not causeneurovirulence in adult WT C57BL/6 mice, the WNV-TXstrain is highly neurovirulent (8, 64). Importantly, in isogenicIKK��/� mice, WNV-MAD disseminated to the spleen at ear-lier times and caused increased clinical disease, although noneof the animals succumbed to lethal infection over the studytime course. Furthermore, IFIT2 expression was not sustainedin the spleen of these animals at later time during WNV-MADinfection. However, early IFIT2 induction likely resultedfrom IRF-3 activation, as IFIT2 is responsive to both IRF-3and IFN-stimulation, which can be differentially regulated indifferent cell types of the spleen. These studies confirmedthe IKK� dependence for STAT1 Ser-708 phosphorylationand sustained IFIT2 expression during virus infection todemonstrate a role for the IKK� pathway of IFIT2 expressionin vivo during WNV infection (see Fig. 6). Viral suppressionof IFIT2 or IKK� signaling may therefore impart replicationfitness for the support of viral spread and tissue dissemina-tion and therefore represents a virulence determinantamong strains of WNV and possibly other pathogenicviruses. The IKK� pathway could therefore prove attractivefor therapeutic strategies aimed at limiting virus replicationand enhancing innate antiviral immunity. Further work todefine this pathway will reveal the nature of IKK� signalingcontrol during the response to IFN.

Acknowledgments—We thank Drs. Stacy Horner, Yueh Ming-Loo,CourtneyWilkins, Helene Liu, and Renee Ireton for critical reading ofthe manuscript and Arjun Rustagi for the generation of IRF-3antibody.

REFERENCES1. Lanciotti, R. S., Roehrig, J. T., Deubel, V., Smith, J., Parker, M., Steele, K.,

Crise, B., Volpe, K. E., Crabtree, M. B., Scherret, J. H., Hall, R. A., MacK-enzie, J. S., Cropp, C. B., Panigrahy, B., Ostlund, E., Schmitt, B.,Malkinson,M., Banet, C., Weissman, J., Komar, N., Savage, H. M., Stone, W., McNa-mara, T., and Gubler, D. J. (1999) Science 286, 2333–2337

2. Smithburn, K. C., Hughes, T. P., Burke, A.W., and Paul, J. H. (1940) J. Trop.Med. Hyg. 20, 471–492

3. Berthet, F. X., Zeller, H. G., Drouet, M. T., Rauzier, J., Digoutte, J. P., andDeubel, V. (1997) J. Gen. Virol. 78, 2293–2297

4. Lanciotti, R. S., Ebel, G. D., Deubel, V., Kerst, A. J., Murri, S., Meyer, R.,Bowen, M., McKinney, N., Morrill, W. E., Crabtree, M. B., Kramer, L. D.,and Roehrig, J. T. (2002) Virology 298, 96–105

5. Lvov,D. K., Butenko,A.M.,Gromashevsky, V. L., Kovtunov, A. I., Prilipov,A. G., Kinney, R., Aristova, V. A., Dzharkenov, A. F., Samokhvalov, E. I.,Savage, H. M., Shchelkanov, M. Y., Galkina, I. V., Deryabin, P. G., Gubler,D. J., Kulikova, L. N., Alkhovsky, S. K., Moskvina, T. M., Zlobina, L. V.,Sadykova, G. K., Shatalov, A.G., Lvov, D.N., Usachev, V. E., andVoronina,A. G. (2004) Arch. Virol. Suppl. 85–96

6. Jia, X. Y., Briese, T., Jordan, I., Rambaut, A., Chi, H. C., Mackenzie, J. S.,Hall, R. A., Scherret, J., and Lipkin, W. I. (1999) Lancet 354, 1971–1972

7. Daffis, S., Samuel, M. A., Suthar, M. S., Keller, B. C., Gale, M., Jr., and

FIGURE 7. A model illustrating that early and late ISGs induction is regulated by multiple STAT1 posttranslational modifications. 1, the canonicalJAK-STAT signaling is activated following type I IFN binding to its receptor, which results in STAT1 Tyr-701 phosphorylation, ISGF3 formation, and its nucleartranslocation. ISGF3 binding to the ISRE element induces transcription of ISGs. 2, chromatin-bound STAT1 can be phosphorylated by MAPK at residue Ser-727,which induces its sumoylation. 3 and 4, nuclear STAT1 is also acetylated by histone acetyltransferase (HAT) CREB-binding protein (CBP), resulting in recruitmentof TCP1, which catalyzes STAT1 tyrosine dephosphorylation. Sumoylated-acetylated STAT1 cycles back to the cytoplasm, and both modifications render STAT1unable to be further tyrosine-phosphorylated. 5 and 6, type I IFN signaling and unknown IFN-stimulated factor(s) activate IKK� phosphorylation of STAT1Ser-708. 7, STAT1 molecules phosphorylated at Ser-708 can enter the nucleus and induce expression of a specific ISG subset. pY, tyrosine phosphorylation; pS,serine phosphorylation; Ac, acetylation; Su, sumoylation.

IKK� Restriction of WNV Infection

44422 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 52 • DECEMBER 30, 2011

Diamond, M. S. (2008) J. Virol. 82, 8465–84758. Keller, B. C., Fredericksen, B. L., Samuel, M. A., Mock, R. E., Mason, P.W.,

Diamond, M. S., and Gale, M., Jr. (2006) J. Virol. 80, 9424–94349. Keller, B. C., Johnson, C. L., Erickson, A. K., and Gale, M., Jr. (2007) Cyto-

kine Growth Factor Rev. 18, 535–54410. Fredericksen, B. L., Keller, B. C., Fornek, J., Katze, M. G., and Gale, M., Jr.

(2008) J. Virol. 82, 609–61611. Loo, Y. M., Fornek, J., Crochet, N., Bajwa, G., Perwitasari, O., Martinez-

Sobrido, L., Akira, S., Gill, M. A., García-Sastre, A., Katze,M. G., andGale,M., Jr. (2008) J. Virol. 82, 335–345

12. Suthar, M. S., Ma, D. Y., Thomas, S., Lund, J. M., Zhang, N., Daffis, S.,Rudensky, A. Y., Bevan,M. J., Clark, E. A., Kaja,M. K., Diamond,M. S., andGale, M., Jr. (2010) PLoS Pathog. 6, e1000757

13. Gale, M., Jr., and Foy, E. M. (2005) Nature 436, 939–94514. Saito, T., and Gale, M., Jr. (2007) Curr. Opin. Immunol. 19, 17–2315. Darnell, J. E., Jr., Kerr, I.M., and Stark,G. R. (1994) Science264, 1415–142116. Hengel, H., Koszinowski, U. H., and Conzelmann, K. K. (2005) Trends

Immunol. 26, 396–40117. Levy, D. E., and García-Sastre, A. (2001) Cytokine Growth Factor Rev. 12,

143–15618. Sen, G. C. (2001) Annu. Rev. Microbiol. 55, 255–28119. Uddin, S., Sassano, A., Deb, D. K., Verma, A., Majchrzak, B., Rahman, A.,

Malik, A. B., Fish, E. N., and Platanias, L. C. (2002) J. Biol. Chem. 277,14408–14416

20. Varinou, L., Ramsauer, K., Karaghiosoff, M., Kolbe, T., Pfeffer, K., Müller,M., and Decker, T. (2003) Immunity. 19, 793–802

21. Tenoever, B. R., Ng, S. L., Chua, M. A., McWhirter, S. M., García-Sastre,A., and Maniatis, T. (2007) Science 315, 1274–1278

22. Platanias, L. C. (2005) Nat. Rev. Immunol. 5, 375–38623. Daffis, S., Szretter, K. J., Schriewer, J., Li, J., Youn, S., Errett, J., Lin, T. Y.,

Schneller, S., Zust, R., Dong,H., Thiel, V., Sen, G. C., Fensterl, V., Klimstra,W. B., Pierson, T. C., Buller, R. M., Gale, M., Jr., Shi, P. Y., and Diamond,M. S. (2010) Nature 468, 452–456

24. Fensterl, V., and Sen, G. C. (2011) J. Interferon Cytokine Res. 31, 71–7825. Sarkar, S. N., and Sen, G. C. (2004) Pharmacol. Ther. 103, 245–25926. Terenzi, F., Hui, D. J., Merrick, W. C., and Sen, G. C. (2006) J. Biol. Chem.

281, 34064–3407127. Hui, D. J., Terenzi, F., Merrick, W. C., and Sen, G. C. (2005) J. Biol. Chem.

280, 3433–344028. Wang, C., Pflugheber, J., Sumpter, R., Jr., Sodora, D. L., Hui, D., Sen, G. C.,

and Gale, M., Jr. (2003) J. Virol. 77, 3898–391229. Daffis, S., Samuel, M. A., Keller, B. C., Gale, M., Jr., and Diamond, M. S.

(2007) PLoS Pathog. 3, e10630. Bluyssen, H. A., Vlietstra, R. J., Faber, P. W., Smit, E. M., Hagemeijer, A.,

and Trapman, J. (1994) Genomics 24, 137–14831. Bandyopadhyay, S. K., Leonard, G. T., Jr., Bandyopadhyay, T., Stark, G. R.,

and Sen, G. C. (1995) J. Biol. Chem. 270, 19624–1962932. Wacher, C., Müller, M., Hofer, M. J., Getts, D. R., Zabaras, R., Ousman,

S. S., Terenzi, F., Sen, G. C., King, N. J., and Campbell, I. L. (2007) J. Virol.81, 860–871

33. Hemmi, H., Takeuchi, O., Sato, S., Yamamoto, M., Kaisho, T., Sanjo, H.,Kawai, T., Hoshino, K., Takeda, K., and Akira, S. (2004) J. Exp. Med. 199,1641–1650

34. Foy, E., Li, K.,Wang, C., Sumpter, R., Jr., Ikeda,M., Lemon, S.M., andGale,M., Jr. (2003) Science 300, 1145–1148

35. Peters, R. T., Liao, S. M., and Maniatis, T. (2000)Mol. Cell 5, 513–52236. Breiman, A., Grandvaux, N., Lin, R., Ottone, C., Akira, S., Yoneyama, M.,

Fujita, T., Hiscott, J., and Meurs, E. F. (2005) J. Virol. 79, 3969–397837. Xie, X., Zhang, D., Zhao, B., Lu, M. K., You, M., Condorelli, G., Wang,

C. Y., and Guan, K. L. (2011) Proc. Nat. Acad. Sci. U.S.A. 108, 6474–647938. Pellegrini, S., John, J., Shearer, M., Kerr, I. M., and Stark, G. R. (1989)Mol.

Cell. Biol. 9, 4605–461239. Sato, M., Suemori, H., Hata, N., Asagiri, M., Ogasawara, K., Nakao, K.,

Nakaya, T., Katsuki, M., Noguchi, S., Tanaka, N., and Taniguchi, T. (2000)Immunity. 13, 539–548

40. Haspel, R. L., Salditt-Georgieff, M., and Darnell, J. E., Jr. (1996) EMBO J.15, 6262–6268

41. McBride, K. M., and Reich, N. C. (2003) Sci. STKE 2003, RE1342. Mowen, K., and David, M. (2000)Mol. Cell. Biol. 20, 7273–728143. Jiang, D., Weidner, J. M., Qing, M., Pan, X. B., Guo, H., Xu, C., Zhang, X.,

Birk, A., Chang, J., Shi, P. Y., Block, T.M., and Guo, J. T. (2010) J. Virol. 84,8332–8341

44. Zhang, Y., Burke, C.W., Ryman, K. D., and Klimstra, W. B. (2007) J. Virol.81, 11246–11255

45. Terenzi, F., Saikia, P., and Sen, G. C. (2008) EMBO J. 27, 3311–332146. Daffis, S., Suthar, M. S., Gale, M., Jr., and Diamond, M. S. (2009) J. Innate

Immun. 1, 435–44547. Daffis, S., Lazear, H. M., Liu, W. J., Audsley, M., Engle, M., Khromykh,

A. A., and Diamond, M. S. (2011) J. Virol. 85, 5664–566848. Fitzgerald, K. A., McWhirter, S. M., Faia, K. L., Rowe, D. C., Latz, E.,

Golenbock, D. T., Coyle, A. J., Liao, S. M., and Maniatis, T. (2003) Nat.Immunol. 4, 491–496

49. tenOever, B. R., Sharma, S., Zou,W., Sun, Q., Grandvaux, N., Julkunen, I.,Hemmi, H., Yamamoto, M., Akira, S., Yeh, W. C., Lin, R., and Hiscott, J.(2004) J. Virol. 78, 10636–10649

50. McWhirter, S. M., Fitzgerald, K. A., Rosains, J., Rowe, D. C., Golenbock,D. T., and Maniatis, T. (2004) Proc. Natl. Acad. Sci. U.S.A. 101, 233–238

51. Maher, S. G., Sheikh, F., Scarzello, A. J., Romero-Weaver, A. L., Baker,D. P., Donnelly, R. P., and Gamero, A. M. (2008) Cancer Biol. Ther. 7,1109–1115

52. Wong, A. H., Tam, N. W., Yang, Y. L., Cuddihy, A. R., Li, S., Kirchhoff, S.,Hauser, H., Decker, T., and Koromilas, A. E. (1997) EMBO J. 16,1291–1304

53. Samuel, M. A., Whitby, K., Keller, B. C., Marri, A., Barchet, W., Williams,B. R., Silverman, R. H., Gale,M., Jr., andDiamond,M. S. (2006) J. Virol. 80,7009–7019

54. Gilfoy, F. D., and Mason, P. W. (2007) J. Virol. 81, 11148–1115855. Katsoulidis, E., Li, Y., Mears, H., and Platanias, L. C. (2005) J. Interferon

Cytokine Res. 25, 749–75656. Uddin, S., Lekmine, F., Sharma, N., Majchrzak, B., Mayer, I., Young, P. R.,

Bokoch, G. M., Fish, E. N., and Platanias, L. C. (2000) J. Biol. Chem. 275,27634–27640

57. Uddin, S.,Majchrzak, B.,Woodson, J., Arunkumar, P., Alsayed, Y., Pine, R.,Young, P. R., Fish, E. N., and Platanias, L. C. (1999) J. Biol. Chem. 274,30127–30131

58. Krämer, O. H., Knauer, S. K., Greiner, G., Jandt, E., Reichardt, S., Gührs,K. H., Stauber, R. H., Böhmer, F. D., and Heinzel, T. (2009)Genes Dev. 23,223–235

59. Krämer, O. H., and Heinzel, T. (2010)Mol. Cell. Endocrinol. 315, 40–4860. Vanhatupa, S., Ungureanu, D., Paakkunainen, M., and Silvennoinen, O.

(2008) Biochem. J. 409, 179–18561. Zimnik, S., Gaestel, M., and Niedenthal, R. (2009) Nucleic Acids Res. 37,

e3062. Cheon, H., and Stark, G. R. (2009) Proc. Natl. Acad. Sci. U.S.A. 106,

9373–937863. Yang, J., and Stark, G. R. (2008) Cell Res. 18, 443–45164. Beasley, D. W., Li, L., Suderman, M. T., and Barrett, A. D. (2002) Virology

296, 17–23

IKK� Restriction of WNV Infection

DECEMBER 30, 2011 • VOLUME 286 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 44423